This report is focused around Lost and Found data using the intakes and outcomes data received for 2019-2021. Its goal is to reflect everything we could learn about L&F from the available data, make sure the numbers we see make sense, and highlight things that would be useful to show but some/all data required for them are missing.
Date range: 2019-01-01 to 2021-12-30
Scroll down or use the table of contents on the left to navigate throughout the document. Most sections contain multiple tabs showing different facets of a data type. Most plots are interactive, meaning they include tooltips and allow hiding and showing parts and zooming in and out. If something went wrong, look for the house icon in the top right corner of each figure to reset.
This section provides an overview of the RTH rate per year divided by species.
This table covers all strays and RTHs. RTH rates shown below are the number of strays with RTO outcome out of all strays. Animals with TNR outcomes are excluded.
When we go over this, let’s make sure we calculate the rate the same way you do, so we would want to make sure what we see makes sense. If these numbers are right, they are slightly lower than the HASS average, which are at about 30% RTH rate (for dogs), and show a small increase in 2021 compared to previous years. Cat RTH rate is similar to other shelters.
| Species | Year | Strays | RTH_Count | RTH_Rate |
|---|---|---|---|---|
| Cat | 2019 | 1143 | 20 | 1.7% |
| Cat | 2020 | 1247 | 28 | 2.2% |
| Cat | 2021 | 1283 | 35 | 2.7% |
| Dog | 2019 | 4000 | 832 | 20.8% |
| Dog | 2020 | 3004 | 585 | 19.5% |
| Dog | 2021 | 3329 | 735 | 22.1% |
| Other | 2019 | 39 | 10 | 25.6% |
| Other | 2020 | 32 | 3 | 9.4% |
| Other | 2021 | 74 | 1 | 1.4% |
This graph compares the RTH rate for dogs coming in from the field and over the counter. Cats are excluded because there are very few of them. It seems that field intakes have consistently higher chances of RTH outcomes compared to OTC animals.
This time series shows the RTH rate per month, to see how the yearly rate breaks down throughout the year. While cat RTH is a bit too low to deduce anything from the variations, the dog RTH rate presents a pattern where January-May show an annual peak, after which there is a decline from June onward until the turn of the year.
Cat intake is highly seasonal, dog intakes have gone down significantly when COVID hit and reached pre-pandemic levels in July before dropping down again at the end of 2021.
The average difference in length of stay (in days) between strays with RTH outcomes and all other strays is shown in the table below – roughly 9 days for dogs and 6 for cats.
That means that every successful RTH saves about 9 days of care on average at Lifeline - Fulton, and field RTH would save an extra day or two on average for RTH from the shelter.
This could translate to pretty significant cost savings at scale – assuming a daily cost of care of $30 per animal, if 250 more dogs were returned home in 2021, it would have saved Lifeline - Fulton about $67,500 in costs of care. This is a fairly simple calculation, but it gets at the magnitude of the potential benefits.
| Species | Outcome | Count | Average_Length_Of_Stay |
|---|---|---|---|
| Cat | Other Outcomes | 1762 | 8.33 days |
| Cat | RTO | 83 | 1.98 days |
| Dog | Other Outcomes | 6372 | 12.43 days |
| Dog | RTO | 2152 | 3.29 days |
The following maps show stray intake and RTH rate by Census tracts to highlight geographical patterns. The first and second tab are similar to previous metrics; the third tab, RTH Gap, shows the number of strays who were not returned home per census tract.
The data in this section includes stray animals for which found addresses were present. Out of all strays in the data (14500), about 500 had a found location of the shelter address, which is a reasonable percentage, but a further 2500 animals had unusable found locations for mapping – primarily street names with no number or intersection. These had to be removed, so the mapping below only shows those animals who did have workable data. ~100 animals were removed because they were found outside Fulton County were also removed for simplicity of mapping. 150 animals that are not dogs or cats were not mapped due to their low numbers.
After this filtering, the data below (number of strays, rate of RTH, RTH gap) is shown for 8253 dogs of which 1761 were RTH.
The next section will show the same maps for 2845 stray cats of which 72 were RTH.
This map excludes census tracts with less than 5 strays, for which the RTH rate might not mean a lot but make the overall map harder to read. It seems like in most areas the RTH rate is fairly consistent, which is one of two common patterns, the other being that areas with high stray intake also tend to have lower RTH rate. The uniform rate suggests people from across the county are equally able to reclaim their pets.
This combines the other two tabs to highlight where most additional RTH potential exists - it shows the number of strays NOT returned to home in each area. As the RTH rate is fairly uniform, it looks a lot like the stray intake map.
Here’s a sneak peak into the top 10 found locations plotted above, to make sure they make sense to you.
| Found.Location | Count |
|---|---|
| 401 CARLTON RD, PALMETTO GA 30268 | 23 |
| 981 HOWELL MILL RD NW, ATLANTA GA 30318 | 22 |
| 3142 OAKCLIFF RD, ATLANTA GA 30311 | 21 |
| 245 AURORA AVE NW, ATLANTA GA 30314 | 18 |
| 1565 MANSELL RD, ALPHARETTA GA 30009 | 16 |
| 4555 WASHINGTON RD, ATLANTA GA 30349 | 12 |
| 5401 OLD NATIONAL HWY, ATLANTA GA 30349 | 12 |
| 780 NEAL ST NW, ATLANTA GA 30318 | 12 |
| 1955 COMPTON DR SE, ATLANTA GA 30315 | 11 |
| 2125 JOSEPH E BOONE BLVD NW, ATLANTA GA 30314 | 11 |
This is similar to the stray intake map above, but for 7023 stray cats. Since only 96 of those were RTH, there is no point in mapping those across town.
Here’s a sneak peak into the top 10 found locations plotted above, to make sure they make sense to you.
| Found.Location | Count |
|---|---|
| 5675 ROSWELL RD , FULTON COUNTY, GA | 14 |
| 509 HARRIS AVE , FULTON COUNTY, GA | 13 |
| 5150 THOMPSON RD , FULTON COUNTY, GA | 11 |
| 600 MARTIN ST SE 30312 GA | 11 |
| 981 HOWELL MILL RD NW, ATLANTA GA 30318 | 11 |
| 1071 HOWELL MILL RD NW, ATLANTA GA 30318 | 10 |
| 3075 HOWELL MILL RD , FULTON COUNTY, GA | 10 |
| 104 REVERE TURN , FULTON COUNTY, GA | 9 |
| 1070 HIDDEN POND LN , FULTON COUNTY, GA | 9 |
| 5675 ROSWELL RD NE, ATLANTA GA 30342 | 9 |
This map shows different demographic information for fulton County.
One example of using both the census data and shelter data is below – there is a negative correlation between stray intakes and median household income (so poorer areas tend to have more intakes). This is a typical pattern in communities across the country.
This section examines animals that had an RTH outcome and both a found location and an outcome address listed to find out how far dogs travel from home when they get lost (and are found).
Out of the 2265 RTH outcomes for strays, almost half had outcome addresses in one of the shelter’s locations (Marietta and Peachtree), so they had to be excluded. Few others were removed for a lack of owner address.
For each dog, the listed intake address and owner addresses were geocoded (using Google’s geolocation service), and then the distance between the two points was calculated. 3 geocoding errors were excluded, as well as about 10 animals for whom the resulting distance was over 100 miles (owners out of state).
Unfortunately, at this stage it became clear that about half the remaining animals had identical crossing and outcome addresses (the distance found was 0). At that point, it seemed like the available data was not reliable enough to reflect all RTH outcomes, so we did not continue with the analysis.
This section uses the values in the Intake Exam ‘treatment subtype’ field to determine chip status upon intake. A chip number was assumed to mean there was a chip upon intake, all ‘no chip’ values indicate the opposite, and there were roughly 3000 animals in total (for all years, all species) for which there was no matching exam record.
Using the DOB and intake date fields, animals younger than 8 weeks at intake were excluded from all analysis below.
There are more dogs and coming in microchipped (16.2%) than cats (5.7%). Cats also had far more unknown values, suggesting they were not scanned .
| Species | Microchip | Count | Percentage |
|---|---|---|---|
| Cat | No | 1069 | 57.9% |
| Cat | Unknown | 671 | 36.3% |
| Cat | Yes | 106 | 5.7% |
| Dog | No | 7203 | 73.1% |
| Dog | Unknown | 1060 | 10.8% |
| Dog | Yes | 1595 | 16.2% |
This comparison is stronger after also making sure animals compared are similar on other characteristics, such as intake condition and age. But to get a first impression, for cats the RTH rate with chips is 21% compared to 1% without one, whereas for dogs, there is a 40% RTH rate for dogs with microchips vs 14% without chips.
The difference is obviously high, but it is worth also thinking about what might make the ‘yes’ category be at 40% as opposed to 100% (since there is presumably an owner), such as owners refusing to reclaim, difficulty paying fees, wrong details on the chip, etc. There are 1000 animals (across 2019-2021) that were identified as having a microchip but were not RTH.
Dogs with an unknown chip status were RTH 36% of times, so closer to the ‘yes’ category.
| Species | Microchip | Strays | RTH_Count | RTH_Rate |
|---|---|---|---|---|
| Cat | No | 1069 | 16 | 1% |
| Cat | Yes | 106 | 22 | 21% |
| Dog | No | 7203 | 1026 | 14% |
| Dog | Yes | 1595 | 636 | 40% |
This map shows the the number of animals coming in without a microchip from each Census tract, which could help prioritize microchipping efforts. When only looking at dogs, the map remains fairly similar, so the map below includes all species. The areas that stand out tend to overlap with those with the higher stray intake.
Found location - as mentioned above, many animals had to be removed from mapping because of unusable found locations – primarily street names with no number or intersection. Using street name and number, a block number, or an intersection of two streets would improve mapping abilities.
Owner/outcome addresses were often the shelter address, which meant we couldn’t really perform a realiable distance analysis.
Thanks for reading through, and we’re looking forward to talking through it and thinking about more ways to make this data useful for you.